Basic layouts for modular assembly systems - a simulation-based comparison :
Grundlayouts für modulare Montagesysteme - ein simulationsbasierter Vergleich. - In: Simulation in Produktion und Logistik 2023, (2023), S. 197-206
The article discusses the challenges posed by increased individualization of products, shorter product life cycles, and external factors on the flexibility of modern production systems. In particular, flexible workshop-oriented manufacturing principles are being implemented to replace or supplement traditional assembly lines, with various terms such as "modular assembly" and "matrix production" etc. used to describe similar concepts. The article presents these concepts under the umbrella term of modular production or assembly systems, which utilize adaptable workstations and autonomous vehicles to transport production orders between stations. The design of such systems is crucial to their performance, with considerations such as task allocation, material supply, and fleet sizing requiring complex interplay. The article compares traditional matrix layouts with alternative options, such as single-lane pathways and non-matrix layouts like honeycomb or star shapes, using simulationbased analysis to evaluate their potential impact on system performance.
https://doi.org/10.22032/dbt.57809
Simulation in Produktion und Logistik 2023 : ASIM Fachtagung : 20. Fachtagung, 13.-15. September 2023, TU Ilmenau. - Ilmenau : Universitätsverlag Ilmenau, 2023. - 1 Online-Ressource (XII, 485 Seiten). - (ASIM-Mitteilung ; Nr. 187)
https://doi.org/10.22032/dbt.57476
Optimization of the design of modular production systems. - In: 2022 Winter Simulation Conference (WSC), (2022), S. 1783-1793
The desire for more flexibility in manufacturing systems, especially when different products or many product variants are manufactured in one production system is leading to a move away from the manufacturing principle of classic line production to more flexible and workshop-oriented production systems, particularly in the automotive industry. One of the challenges in these so-called modular assembly or production systems is the system design, especially the allocation of activities to the individual production cells. One approach to improve this allocation is offered by simulation-based optimization. In this paper, a concept for simulation-based optimization of the design of modular production systems is presented and demonstrated by means of a small academic case study. Classical genetic algorithms and additionally the NSGA-II algorithm, which also allows multi-objective optimization, are used.
https://doi.org/10.1109/WSC57314.2022.10015350
Development of an integrated solution for data farming and knowledge discovery in simulation data. - In: Simulation Notes Europe, ISSN 2164-5353, Bd. 32 (2022), 3, S. 121-126
Simulation is an established methodology for planning and evaluating manufacturing and logistics systems. In contrast to classical simulation studies, the method of knowledge discovery in simulation data uses a simulation model as a data generator (data farming). Subsequently, hidden, previously unknown and potentially useful cause-effect relationships can be uncovered on the generated data using data mining and visual analytics methods. So far, however, there was a lack of integrated, easy-to-use software solutions for the application of the data farming in operational practice. This paper presents such an integrated solution, which allows generating experiment designs, implements a method to distribute the necessary experiment runs, and provides the user with tools to analyze and visualize the result data.
https://dx.doi.org/10.11128/sne.32.tn.10611
Simulationsbasiertes Deep Reinforcement Learning für Modulare Produktionssysteme. - In: ASIM SST 2022 Proceedings Langbeiträge, (2022), S. 65-72
Modulare Produktionssysteme sollen die traditionelle Fließbandproduktion in der Automobilindustrie ablösen. Die Idee dabei ist, dass sich hochgradig individualisierte Produkte dynamisch und autonom durch ein System flexibler Arbeitsstationen bewegen können. Dieser Ansatz stellt hohe Anforderungen an die Planung und Organisation solcher Systeme. Da jedes Produkt seinen Weg durch das System frei und individuell bestimmen kann, kann die Implementierung von Regeln und Heuristiken, die die Flexibilität des Systems zur Leistungssteigerung ausnutzen, in diesem dynamischen Umfeld schwierig sein. Transportaufgaben werden in der Regel von fahrerlosen Transportsystemen (FTS) ausgeführt. Daher bietet die Integration von KI-basierten Steuerungslogiken eine vielversprechende Alternative zu manuell implementierten Entscheidungsregeln für den Betrieb der FTS. In diesem Beitrag wird ein Ansatz für den Einsatz von Reinforcement Learning (RL) in Kombination mit Simulation vorgestellt, um FTS in modularen Produktionssystemen zu steuern. Darüber hinaus werden Untersuchungen zu dessen Flexibilität und Skalierbarkeit durchgeführt.
https://dx.doi.org/10.11128/arep.20.a2007
Optimierung des Designs modularer Montagesysteme. - In: ASIM SST 2022 Proceedings Langbeiträge, (2022), S. 15-22
Der Wunsch nach mehr Flexibilität in Fertigungssystemen, insbesondere, wenn verschiedene Produkte bzw. viele Produktvarianten in einem Produktionssystem gefertigt werden, führt, besonders in der Automobilindustrie, zur Abkehr vom Fertigungsprinzip der klassischen Linienfertigungen hin zu eher flexiblen und werkstattorientierten Produktionssystemen. Eine der Herausforderungen in diesen so genannten modularen Montage- bzw. Produktionssystemen ist das Systemdesign, insbesondere die Zuordnung der Tätigkeiten auf die einzelnen Fertigungsinseln. Ein Ansatz, diese Zuordnung zu verbessern bietet die simulationsbasierte Optimierung. In diesem Beitrag wird ein Konzept zur simulationsbasierten Optimierung des Designs modularer Montagesysteme vorgestellt und anhand einer Fallstudie demonstriert. Zum Einsatz kommen hierbei genetische Algorithmen, speziell der NSGA-II-Algorithmus, welcher auch mehrkriterielle Optimierung ermöglicht.
https://dx.doi.org/10.11128/arep.20.a2006
Data farming and knowledge discovery in simulation data : development of an integrated solution
Data Farming und Wissensentdeckung in Simulationsdaten : Entwicklung einer integrierten Lösung. - In: Zeitschrift für wirtschaftlichen Fabrikbetrieb, ISSN 2511-0896, Bd. 117 (2022), 3, S. 144-150
Simulation als Methode der Digitalen Fabrik hat sich seit langem zur Unterstützung der Planung von Produktions- und Logistiksystemen etabliert. In Ergänzung zu bisher vorherrschenden Simulationsstudien wird bei der hier vorgestellten Methode der Wissensentdeckung in Simulationsdaten ein Simulationsmodell als Datengenerator verwendet. Dadurch können mittels Data-Mining- und Visual-Analytics-Methoden versteckte und potenziell nützliche Ursache-Wirkungs-Beziehungen in den generierten Daten aufgedeckt werden. Bislang fehlte es jedoch an integrierten Softwarelösungen für die Praxis.
https://doi.org/10.1515/zwf-2022-1032
A method using generative adversarial networks for robustness optimization. - In: ACM transactions on modeling and computer simulation, ISSN 1558-1195, Bd. 32 (2022), 2, S. 12:1-12:22
The evaluation of robustness is an important goal within simulation-based analysis, especially in production and logistics systems. Robustness refers to setting controllable factors of a system in such a way that variance in the uncontrollable factors (noise) has minimal effect on a given output. In this paper, we present an approach for optimizing robustness based on deep generative models, a special method of deep learning. We propose a method consisting of two Generative Adversarial Networks (GANs) to generate optimized experiment plans for the decision factors and the noise factors in a competitive, turn-based game. In a case study, the proposed method is tested and compared to traditional methods for robustness analysis including Taguchi method and Response Surface Method.
https://doi.org/10.1145/3503511
Automation within the process of knowledge discovery in simulation data : characterization of the result data
Automatisierung im Prozess der Wissensentdeckung in Simulationsdaten : Charakterisierung der Ergebnisdaten. - In: Simulation in Produktion und Logistik 2021, (2021), S. 367-376
Literaturangaben
The traditional application of simulation in production and logistics is usually aimed at changing certain parameters in order to answer clearly defined objectives or questions. In contrast to this approach, the method of knowledge discovery in simulation data (KDS) uses a simulation model as a data generator (data farming). Subsequently using data mining methods, hidden, previously unknown and potentially useful cause-effect relationships can be uncovered. So far, however, there is a lack of guidelines and automatization-tools for non-experts or novices in KDS, which leads to a more difficult use in industrial applications and prevents a broader utilization. This paper presents a concept for automating the first step of the KDS, which is the process of characterization of the result data, using meta learning and validates it on small case study.
Development of an integrated solution for data farming and knowledge discovery in simulation data :
Entwicklung einer integrierten Lösung für das Data Farming und die Wissensentdeckung in Simulationsdaten. - In: Simulation in Produktion und Logistik 2021, (2021), S. 377-386
Literaturangaben
Simulation is an established methodology for planning and evaluating manufacturing and logistics systems. In contrast to classical simulation studies, the method of knowledge discovery in simulation data uses a simulation model as a data generator (data farming). Subsequently, hidden, previously unknown and potentially useful cause-effect relationships can be uncovered on the generated data using data mining and visual analytics methods. So far, however, there is a lack of integrated, easy-to-use software solutions for the application of the data farming in operational practice. This paper presents such an integrated solution, which allows for generating experiment designs, implements a method to distribute the necessary experiment runs, and provides the user with tools to analyze and visualize the result data.
Simulation-based deep reinforcement learning for modular production systems. - In: 2020 Winter Simulation Conference (WSC), (2020), S. 1596-1607
Modular production systems aim to supersede the traditional line production in the automobile industry. The idea here is that highly customized products can move dynamically and autonomously through a system of flexible workstations without fixed production cycles. This approach has challenging demands regarding planning and organization of such systems. Since each product can define its way through the system freely and individually, implementing rules and heuristics that leverage the flexibility in the system in order to increase performance can be difficult in this dynamic environment. Transport tasks are usually carried out by automated guided vehicles (AGVs). Therefore, integration of AI-based control logics offer a promising alternative to manually implemented decision rules for operating the AGVs. This paper presents an approach for using reinforcement learning (RL) in combination with simulation in order to control AGVs in modular production systems. We present a case study and compare our approach to heuristic rules.
https://doi.org/10.1109/WSC48552.2020.9384089
Knowledge discovery in simulation data. - In: ACM transactions on modeling and computer simulation, ISSN 1558-1195, Bd. 30 (2020), 4, S. 24:1-24:25
This article provides a comprehensive and in-depth overview of our work on knowledge discovery in simulations. Application-wise, we focus on manufacturing simulations. Specifically, we propose and discuss a methodology for designing, executing, and analyzing large-scale simulation experiments with a broad coverage of possible system behavior targeted at generating knowledge about the system. Based on the concept of data farming, we suggest a two-phase process which starts with a data generation phase, in which a smart experiment design is used to set up and efficiently execute a large number of simulation experiments. In the second phase, the knowledge discovery phase, data mining and visually aided analysis methods are applied on the gathered simulation input and output data. This article gives insights into this knowledge discovery phase by discussing different machine learning approaches and their suitability for different manufacturing simulation problems. With this, we provide guidelines on how to conduct knowledge discovery studies within the manufacturing simulation context. We also introduce different case studies, both academic and applied, and use them to validate our methodology.
https://doi.org/10.1145/3391299
A method for robustness optimization using generative adversarial networks. - In: SIGSIM-PADS '20, (2020), S. 1-10
This paper presents an approach for optimizing the robustness of production and logistic systems based on deep generative models, a special method of deep learning. Robustness here refers to setting controllable factors of a system in such a way that variance in the uncontrollable factors (noise) has a minimal effect on given output parameters. In a case study, the proposed method is tested and compared to a traditional method for robustness analysis. The basic idea is to use deep neural networks to generate data for experiment plans and rate them by use of a simulation model of the production system. We propose to use two Generative Adversarial Networks (GANs) to generate optimized experiment plans for the decision factors and the noise factors, respectively, in a competitive, turn-based game. In one turn, the controllable factors are optimized and the noise remains constant, and vice versa in the next turn. For the calculations of the robustness, the planned experiments are conducted and rated using a simulation model in each learning step.
https://doi.org/10.1145/3384441.3395981
Automatische Modellgenerierung - Stand, Klassifizierung und ein Anwendungsbeispiel. - In: Ablaufsimulation in der Automobilindustrie, (2020), S. 333-347
Die automatische Modellgenerierung (AMG) ist ein Ansatz, der darauf abzielt, sowohl die Aufwände einer Simulationsstudie zu senken als auch die Qualität der erzeugten Modelle zu verbessern. Unter automatischer Modellgenerierung werden im Kontext der Simulation verschiedene Ansätze subsumiert, die es erlauben, Simulationsmodelle oder zumindest Teile von Simulationsmodellen mittels Algorithmen zu erzeugen. Eine umfassende Klassifizierung der Ansätze nach verschiedenen Merkmalen ist Ausgangspunkt weiterer Betrachtungen des Beitrags, in denen u. a. verschiedene technische Ansätze zur Modellgenerierung diskutiert werden. Weiterhin werden ergänzende Techniken, die die eigentliche Modellgenerierung flankierenden, wie z. B. die automatische Modellinitialisierung, diskutiert. Als ein möglicher Lösungsansatz wird beispielhaft ein Framework zur automatischen Modellgenerierung, -initialisierung und -adaption, welches das standardisierte Core Manufacturing Simulation Data (CMSD) Format als Basis nutzt, beschrieben.
https://doi.org/10.1007/978-3-662-59388-2_23
Visualization and interaction for knowledge discovery in simulation data. - In: Hawaii International Conference on System Sciences 2020, (2020), S. 1340-1349
Discrete-event simulation is an established and popular technology for investigating the dynamic behavior of complex manufacturing and logistics systems. Besides traditional simulation studies that focus on single model aspects, data farming describes an approach for using the simulation model as a data generator for broad scale experimentation with a broader coverage of the system behavior. On top of that we developed a process called knowledge discovery in simulation data that enhances the data farming concept by using data mining methods for the data analysis. In order to uncover patterns and causal relationships in the model, a visually guided analysis then enables an exploratory data analysis. While our previous work mainly focused on the application of suitable data mining methods, we address suitable visualization and interaction methods in this paper. We present those in a conceptual framework followed by an exemplary demonstration in an academic case study.
https://doi.org/10.24251/HICSS.2020.165
Modelling and simulation of modular production systems :
Modellierung und Simulation von modularen Produktionssystemen. - In: Simulation in Produktion und Logistik 2019, (2019), S. 391-401
Modular production systems aim to supersede the traditional line production in the automobile industry. The idea here is that highly customized products can move dynamically and autonomously through a system of flexible workstations without fixed production cycles. This approach has challenging demands regarding planning and organization of such systems. The use of modelling and simulation methods is therefore indispensable. This paper presents simulation approaches for modelling modular production systems and discusses a comparison between an agent-based and a process-oriented implementation of an example model.
Deep learning based prediction of energy consumption for hybrid simulation :
Deep-Learning-basierte Prognose von Stromverbrauch für die hybride Simulation. - In: Simulation in Produktion und Logistik 2019, (2019), S. 121-131
Modern production facilities need to prepare for changing market conditions within the energy market due to ongoing implementation of governmental policies. This results in higher volatility of the availability of energy and therefore energy costs. If a simulation model of a machinery model can estimate its own future consumption, and according time frames for said consumption, this information could be used for optimized scheduling of energy consuming jobs. This would result in lower procurement costs. To make said estimation about the dynamic behaviour of jobs, methods of time series prediction tend to be applied. Here a proposal is made to apply a Hybrid System Model incorporating a recurrent neural network (RNN)-Encoder-Decoder-Architecture, which returns a discrete times series when a behavioural sequence (such as an NC-Code) has been put into a neural net model of the respective machinery. Those discrete time series reflect the machines energy consumption for each job that it has been operated on. This neural net, if weighted and called, emits the length value of a job and an according time series which displays the quasi-continuous time consumption of said job. Such generative models combined with classic simulation paradigm qualify as potent applications of hybrid simulation approaches.
Knowledge discovery and robustness analysis for simulation models of global networks :
Wissensentdeckung und Robustheitsanalyse für Simulationsmodelle weltweiter Netze, (2019), S. 64-76
http://ceur-ws.org/Vol-2397/paper9.pdf
Knowledge discovery in scheduling systems using evolutionary bilevel optimization and visual analytics. - In: Evolutionary multi-criterion optimization, (2019), S. 439-450
https://doi.org/10.1007/978-3-030-12598-1_35
Combining data farming and data envelopment analysis for measuring productive efficiency in manufacturing simulations. - In: Simulation for a noble cause, (2018), S. 1440-1451
Discrete event simulation is an established methodology for investigating the dynamic behavior of complex manufacturing and logistics systems. In addition to traditional simulation studies, the concept of data farming and knowledge discovery in simulation data is a current research topic that consist of broad scale experimentation and data mining assisted analysis of massive simulation output data. While most of the current research aims to investigate key drivers of production performance, in this paper we propose a methodology for investigating productive efficiency. We therefore developed a concept of combining our existing approach of data farming and visual analytics with data envelopment analysis (DEA), which is used to investigate efficiency in operations research and economics. With this combination of concepts, we are not only able to determine key factors and interactions that drive productive efficiency in the modeled manufacturing system, but also to identify the most productive settings.
https://doi.org/10.1109/WSC.2018.8632300
Data Farming und simulationsbasierte Robustheitsanalyse für Fertigungssysteme. - In: ASIM 2018 - 24. Symposium Simulationstechnik, (2018), S. 243-252
Diskrete Simulation ist eine etablierte Methodik zur Untersuchung des dynamischen Verhaltens von komplexen Fertigungs- und Logistiksystemen. Konventionelle Simulationsstudien fokussieren auf einzelne Modellaspekte und spezifische Analysefragen. Der Umfang der ausgeführten Szenarien ist häufig gering. Das Konzept des Data-Farming verwendet das Simulationsmodell als Datengenerator für eine breite Skale von Experimenten und ermöglicht unter Nutzung von Data-Mining-Methoden eine wesentlich breitere Untersuchung des simulierten Systems sowie eine höhere Komplexität in den abgeleiteten Erkenntnissen. Anforderungen an Simulationssysteme und -modelle zur Durchführung von Data-Farming werden erläutert. Eine Erweiterung des Ansatzes ist die simulationsbasierte Robustheitsanalyse auf der Basis von Verlustfunktionen nach Taguchi. Beide Vorgehensweisen werden an einer Fallstudie aus dem Fahrzeugbau demonstriert.
Bilevel innovization: knowledge discovery in scheduling systems using evolutionary bilevel optimization and visual analytics. - In: GECCO'18 companion, ISBN 978-1-4503-5764-7, (2018), S. 197-198
https://doi.org/10.1145/3205651.3205726
Online analysis of simulation data with stream-based data mining. - In: SIGSIM-PADS'17, (2017), S. 241-248
https://doi.org/10.1145/3064911.3064915
Knowledge discovery in simulation data - a case study for a backhoe assembly line. - In: WSC'17, ISBN 978-1-5386-3428-8, (2017), S. 4456-4458
Discrete event simulation is an established and popular technology for investigating the dynamic behavior of complex manufacturing and logistics systems. Besides conventional simulation studies that focus on single model aspects answering project specific analysis questions, new methods of broad scale experiment design and system analysis emerge alongside new developments of computational power and data processing. This enables to investigate the bandwidth of possible system behavior in a more in-depth way. In this work we applied our previously developed methodology of knowledge discovery in simulation data onto an industrial case study for a backhoe loader manufacturing facility.
https://doi.org/10.1109/WSC.2017.8248162
Knowledge discovery and robustness analysis in manufacturing simulations. - In: WSC'17, ISBN 978-1-5386-3428-8, (2017), S. 3952-3963
Discrete event simulation is an established methodology for investigating the dynamic behavior of complex manufacturing and logistics systems. Traditionally, simulation experts conduct experiments for predetermined system specifications focusing on single model aspects and specific analysis questions. In addition to that, the concept of data farming and knowledge discovery is an ongoing research issue that consists of broad scale experimentation and data mining assisted analysis of massive simulation output data. As an extension to this approach, we propose a concept for investigating the robustness of complex manufacturing and logistic systems which are often very sensitive to variation and noise. Based on Taguchis loss function, we developed a concept including data farming and visual analytics methodologies to investigate sources of variation in a model and the factor values that make a configuration robust. The concept is demonstrated on an exemplary case study model.
https://doi.org/10.1109/WSC.2017.8248105
Data farming for production and logistics :
Data Farming im Kontext von Produktion und Logistik. - In: Simulation in Produktion und Logistik 2017, ISBN 978-3-7376-0192-4, (2017), S. 169-178
Simulation is an established methodology for planning and evaluating manufacturing and logistics systems. Usually simulations experts conduct experiments for a prior defined goal and by minimizing the number of simulation runs. In contrast to that, data farming describes an approach for using the simulation model as a data generator for broad scale experimentation with a broader coverage of system behaviour. This paper demonstrates how to apply data farming methodologies on simulation models in the context of production and logistics and how to analyse massive amounts of simulation data using data mining and visual analytics.
Emulation of control strategies through machine learning in manufacturing simulations. - In: Journal of simulation, ISSN 1747-7786, Bd. 11 (2017), 1, S. 38-50
Discrete-event simulation is a well-accepted method for planning, evaluating, and monitoring processes in production and logistics. To reduce time and effort spent on creating simulation models, automatic simulation model generation is an important area in modeling methodology research. When automatically generating a simulation model from existing data sources, the correct reproduction of dynamic behavior of the modeled system is a common challenge. One example is the representation of dispatching and scheduling strategies of production jobs. When generating a model automatically, the underlying rules for these strategies are typically unknown but yet have to be adequately emulated. In this paper, we summarize our work investigating the suitability of various data mining and supervised machine learning methods for emulating job scheduling decisions based on data obtained from production data acquisition. We report on the performance of the algorithms and give recommendations for their application, including suggestions for their integration in simulation systems.
http://dx.doi.org/10.1057/s41273-016-0006-0
Knowledge discovery in simulation data: a case study of a gold mining facility. - In: Simulating complex service systems, ISBN 978-1-5090-4486-3, (2016), S. 1607-1618
Discrete event simulation is an established methodology for investigating the dynamic behavior of complex systems. Apart from conventional simulation studies, which focus on single model aspects and answering specific analysis questions, new methods of broad scale experiment design and analysis emerge in alignment with new possibilities of computation and data processing. This paper outlines a visually aided process for knowledge discovery in simulation data which is applied onto a real world case study of a mining facility in Western Australia.
https://doi.org/10.1109/WSC.2016.7822210
Innovative Analyse- und Visualisierungsmethoden für Simulationsdaten. - In: , (2016), S. 1737-1748
https://nbn-resolving.org/urn:nbn:de:gbv:ilm1-2016100035
Gestaltungsmöglichkeiten selbst-adaptierender Simulationsmodelle. - In: , (2016), S. 1713-1724
https://nbn-resolving.org/urn:nbn:de:gbv:ilm1-2016100035
Approximation of dispatching rules for manufacturing simulation using data mining methods. - In: Proceedings of the 2015 Winter Simulation Conference, ISBN 978-1-4673-9743-8, (2015), S. 2329-2340
Discrete-event simulation is a well-accepted method for planning, evaluating, and monitoring processes in production and logistics contexts. In order to reduce time and effort spent on creating the simulation model, automatic simulation model generation is an important area in modeling methodology research. When automatically generating a simulation model from existing data sources, the correct reproduction of the dynamic behavior of the modelled system is a common challenge. One example is the representation of dispatching and scheduling strategies of production jobs. When generating a model automatically, the underlying rules for these strategies are typically unknown but yet have to be adequately emulated. In previous work, we presented an approach to approximate the behavior through artificial neural networks. In this paper, we investigate the suitability of various other data mining and supervised machine learning methods for emulating job scheduling decisions with data obtained from production data acquisition.
http://dx.doi.org/10.1109/WSC.2015.7408344
Visual analytics of manufacturing simulation data. - In: Proceedings of the 2015 Winter Simulation Conference, ISBN 978-1-4673-9743-8, (2015), S. 779-790
Discrete event simulation is an accepted technology for investigating the dynamic behavior of complex manufacturing systems. Visualizations created within simulation studies often focus on the animation of the dynamic processes of a single simulation run, supplemented with graphs of certain performance indicators obtained from replications of a simulation run or a few manually conducted simulation experiments. This paper suggests a much broader visually aided analysis of simulation input and output data and their relations than it is commonly applied today. Inspired from the idea of visual analytics, we suggest the application of data farming approaches for obtaining datasets of a much broader spectrum of combinations of input and output data. These datasets are then processed by data mining methods and visually analyzed by the simulation experts. This process can uncover causal relationships in the model behavior that were previously not known, leading to a better understanding of the systems behavior.
http://dx.doi.org/10.1109/WSC.2015.7408215
On the use of the Core Manufacturing Simulation Data (CMSD) standard: experiences and recommendations. - Ilmenau : Univ.-Bibliothek. - Online-Ressource (PDF-Datei: 11 S.)Druck-Ausgabe: Fall Simulation Interoperability Workshop (2015 Fall SIW) : Orlando, Florida, USA, 31 August-4 September 2015 / SISO - Simulation Interoperability Standards Organization. - Red Hook, NY : Curran Associates, Inc., 2016. - Seite 119-129
The Core Manufacturing Simulation Data (CMSD) information model is defined by SISO standards SISO-STD-008-01-2012 and SISO-STD-008-2010. The main objective of CMSD is to facilitate interoperability between simulation systems and other information systems in the manufacturing domain. While CMSD is mainly intended as standardized data exchange format, its capabilities go beyond simple data exchange. Frequently CMSD based system descriptions are used for purposes of automatic simulation model generation. In this paper, we report on practical experiences using the CMSD standard for such purposes as well as for purposes of simulation model initialization and simulation output data collection. Based on our experiences we suggest potential enhancements for a future revision of the standard.
http://www.db-thueringen.de/servlets/DocumentServlet?id=26816
Emulation of control strategies through machine learning in manufacturing simulations :
Abbildung von Steuerungslogiken durch maschinelles Lernen für die Simulation von Produktionssystemen. - In: Simulation in production and logistics 2015, (2015), S. 481-490
In the context of discrete-event simulation of production and logistics systems, modelling an exact representation of the real system is needed for obtaining sound and reliable results. The automatic generation of simulation models is an approach for saving time and effort for creating models and, therefore, it is a recurring issue in modelling methodology research. In automatic model generation, the approximation of dynamic behaviour is a challenging problem. This is for example the case when the dispatching and scheduling of production jobs needs to be adequately emulated, but the underlying rules are unknown. In previous work, we presented an approach for approximating dynamic behaviour through artificial neural networks. In this paper, we propose an improved approach and investigate its suitability again with artificial neuronal networks as well as with other data mining and supervised machine learning methods.
Knowledge discovery in manufacturing simulations. - In: SIGSIM PADS'15, ISBN 978-1-4503-3565-2, (2015), S. 3-12
Discrete event simulation studies in a manufacturing context are a powerful instrument when modeling and evaluating processes of various industries. Usually simulation experts conduct simulation experiments for a predetermined system specification by manually varying parameters through educated assumptions and according to a prior defined goal. Moreover, simulation experts try to reduce complexity and number of simulation runs by excluding parameters that they consider as not influential regarding the simulation project scope. On the other hand, today's world of big data technology enables us to handle huge amounts of data. We therefore investigate the potential benefits of designing large scale experiments with a much broader coverage of possible system behavior. In this paper, we propose an approach for applying data mining methods on simulation data in combination with suitable visualization methods in order to uncover relationships in model behavior to discover knowledge that otherwise would have remained hidden. For a prototypical demonstration we used a clustering algorithm to divide large amounts of simulation output datasets into groups of similar performance values and depict those groups through visualizations to conduct a visual investigation process of the simulation data.
Webbasierte Animation von Simulationsläufen auf Basis des Core Manufacturing Simulation Data (CMSD) Standards. - In: ASIM 2014, 22. Symposium Simulationstechnik, 3. bis 5. September 2014, HTW Berlin; Tagungsband, (2014), S. 63-70
Animation von Simulationsläufen ist für viele Anwendungen ein nicht zu unterschätzendes Hilfsmittel. Die Nutzungsmöglichkeiten sind hierbei mannigfaltig, sie reichen von der Validierung der Modelle bis hin zur Ergebnispräsentation von Simulationsstudien. Dem Nutzen steht mitunter aber auch ein nicht zu unterschätzender Aufwand gegenüber, gerade im Kontext der automatischen webbasierten Simulation sind oft geeignete Animationen nicht verfügbar. Im Rahmen dieses Papers wird ein Ansatz vorgestellt, welcher ein bestehendes Framework zur automatischen Modellgenerierung, -initialisierung und Simulationsdurchführung inklusive Ergebnisauswertung auf Basis des Core Manufacturing Simulation Data (CMSD) Standards um die Möglichkeit der vollständig automatischen webbasierten Animation erweitert. Hierzu wird neben der Diskussion der Grundlagen der Animation das bestehende Framework und der dem Framework zugrunde liegende CMSD-Standard vorgestellt. Des Weiteren werden verschiedene Implementierungstechnologien vom Streamen von Videos über das Nutzen von Plug-Ins wie Flash oder Java Applets bis hin zu modernen Techniken wie HTML 5, CSS3 und JavaScript kritisch beleuchtet. Abschließend wird eine prototypische Implementierung mittels HTML 5 Canvas und den JavaScript Frameworks JQuery und KineticJS vorgestellt.
On the use of artificial neural networks in simulation-based manufacturing control. - In: Journal of simulation, ISSN 1747-7786, Bd. 8 (2014), 1, S. 76-90
The automatic generation of simulation models has been a recurring research topic for several years. In manufacturing industries, it is currently also becoming a topic of high practical relevance. A well-known challenge in most model generation approaches is the correct reproduction of the dynamic behaviour of model elements, for example, buffering or control strategies. This problem is especially relevant in simulation-based manufacturing control. In such scenarios, simulation models need to reflect the current state and behaviour of the real system in a highly accurate way, otherwise its suggested control decisions may be inaccurate or even dangerous towards production goals. This paper introduces a novel methodology for approximating dynamic behaviour using artificial neural networks, rather than trying to determine exact representations. We suggest using neural networks in conjunction with traditional material flow simulation systems whenever a certain decision cannot be made ex ante in the model generation process due to insufficient knowledge about the behaviour of the real system. In such cases the decision is delegated to the neural network, which is connected to the simulation system at runtime. Training of the neural network is performed by observation of the real systems decision and based on the evaluation of data that can be gained through production data acquisition. Our approach has certain advantages compared to other approaches and is especially well suited in the context of on-line simulation and simulation-based operational decision support. We demonstrate the applicability of our methodology using a case study and report on performance results.
http://dx.doi.org/10.1057/jos.2013.6
Automatische Generierung adaptiver Modelle zur Simulation von Produktionssystemen. - Ilmenau : Universitätsverlag Ilmenau, 2013. - Online-Ressource : Ilmenau, Techn. Univ., Diss., 2013
Parallel als Druckausg. erschienen
Die Simulation von Produktionsprozessen wird heute in einer Vielzahl von Branchen eingesetzt. Simulation dient hierbei zur Analyse, dem Design und der Optimierung der Produktions- und Logistikprozesse und dem dabei anfallenden Ressourceneinsatz und kann hierbei sowohl in der Planung, Inbetriebnahme als auch während des operativen Betriebs genutzt werden. Den unbestritten großen Potentialen der Materialflusssimulation in Unternehmen stehen entsprechend hohe Aufwände entgegen. Diese entstehen sowohl bei der Implementierung der Modelle als auch deren Nutzung. Durch schlechte Integration und Standardisierung der Simulation, steigende Anforderungen der Unternehmen bzgl. Genauigkeit, Flexibilität, Anpassbarkeit, Geschwindigkeit, Kosten, Wiederverwendbarkeit, Zyklen und phasenübergreifender Nutzbarkeit usw. werden die Aufwände teils unnötig gesteigert. Ein Ansatz, der seit einigen Jahren immer wieder als ein Lösungsbeitrag für eine bessere Nutzung der Simulation auch gerade in KMU's betrachtet wird, ist die automatische Generierung von Simulationsmodellen. Unter automatischer Modellgenerierung werden verschiedene Ansätze subsumiert, die erlauben Simulationsmodelle oder zumindest Teile von Modellen mittels Algorithmen zu erzeugen. Bisher wurde kein Ansatz veröffentlicht, der für einen breiteren Nutzerkreis und über einen speziellen Teilbereich hinaus gute Ergebnisse liefert.In dieser Arbeit wurde ein umfassendes Rahmenwerk zur Integration bzw. Automatisierung der Simulation entworfen und validiert. Es wurden sowohl organisatorische, methodische als auch prototypisch technische Komponenten betrachtet. In diesem Zusammenhang wird die These vertreten, dass eine breit anwendbare automatische Modellgenerierung allein durch die Nutzung von Standards zum Datenaustausch bzw. zur Konzeptmodellerstellung sinnvoll zu implementieren ist. Konkret wurde der Core Manufacturing Simulation Data (CMSD) Standard genutzt bzw. bildet die Referenzanwendung des Standards die Basis der gesamten Arbeit. Die Unterstützung aller Simulationsphasen, d.h. nicht allein der Modellerstellung sondern auch der Alternativenbildung, Initialisierung, Ergebnisauswertung usw. in allen Komponenten wird durchgehend gewährleistet. Weiterhin wurden konkret Modellgenerierungsmethoden und Verfahren zur Abbildung des dynamischen Verhaltens in Modellen klassifiziert und einzelne Lösungsansätze vorgestellt.
http://www.db-thueringen.de/servlets/DocumentServlet?id=23106
A new web based method for distribution of simulation experiments based on the CMSD standard. - In: Proceedings of the 2012 Winter Simulation Conference (WSC), ISBN 978-1-4673-4779-2, (2012), insges. 12 S.
This article introduces a novel methodology for web based distribution of simulation experiments. The approach is related to themes such as web based applications, cloud computing or applications as a service, which have been recurring topics in scientific papers for years. The methodology is based on automatic model generation, initialization, and result analysis under usage of the CMSD standard. All user interactions are performed in web based user interfaces. Of special importance is that different simulations tools can be used in parallel without any additional effort. Furthermore the simulation tool actually used is transparent to the user. The applicability of our methodology is demonstrated for different production scenarios.
http://dx.doi.org/10.1109/WSC.2012.6464985
Model generation in SLX using CMSD and XML stylesheet transformations. - In: Proceedings of the 2012 Winter Simulation Conference (WSC), ISBN 978-1-4673-4779-2, (2012), insges. 11 S.
This article introduces a novel methodology for automatic simulation model generation. The methodology is based on the usage of XML stylesheet transformations for generating the actual source code of the target simulation system. It is therefore especially well-suited for all language-based simulation systems. The prerequisite for using the methodology is an appropriate representation of the system under investigation in the Core Manufacturing Simulation Data (CSMD) format. The applicability of our methodology is demonstrated for the simulation language SLX as well as for the visualization system Proof Animation.
http://dx.doi.org/10.1109/WSC.2012.6464981
Generation of alternatives for model predictive control in manufacturing. - In: I3M 2012 conference proceedings, (2012), S. 7-16
Manufacturing systems are dynamic systems which are influenced by various disturbances or frequently changing customer requests. A continuous process of decision making is required. Model Predictive Control is a common model-based approach for control but needs adaption to be applicable to discrete-event simulation. In this paper we introduce an approach to model and generate non trivial control options and decisions often made in the operation of manufacturing systems. We also show how complex scenarios can be generated. To support a wide-range of applications our approach is based on the core manufacturing simulation data (CMSD) information model. We implement the design and generation of complex scenarios by processing and combining modeled control options. By using our approach, which also applicable to decision support systems, we can enable model-based closed-loop control based on a symbiotic simulation system and automated model generation and initialization.
Initialization of simulation models using CMSD. - In: Proceedings of the 2011 Winter Simulation Conference, ISBN 978-1-457-72106-9, (2011), S. 2228-2239
In the context of online- and symbiotic simulation, the precise initialization of simulation models based on the state of the physical system is a fundamental requirement. In these simulations, the simulation model typically serves as an operational decision support tool. Obviously, it can therefore not start empty and idle. The accurate capturing of initial conditions is fundamental for the quality of the model based predictions. In literature, it is only generally stated that the simulation model must maintain a close connection with the physical system. Our work systematically investigates which data from the physical system is needed for initialization, how it shall be transferred into the simulation model in a standardized way, and which potential problems must be solved in the simulation system to adequately initialize its model elements. We present a solution based on the core manufacturing simulation data (CMSD) standard, suggest necessary extensions and demonstrate a prototypical implementation.
Automatische Generierung und Initialisierung von Simulationsmodellen unter Verwendung des Core Manufacturing Simulation Data (CMSD) Information Model. - In: Nachhaltigkeit in Fabrikplanung und Fabrikbetrieb, (2011), S. 323-331
Simulation hat sich in vielen Anwendungsbereichen innerhalb produzierender Unternehmen zu einem nicht mehr weg zu denkenden Werkzeug zur Entscheidungsfindung bzw. -unterstützung für nachhaltiges Wirtschaften entwickelt. Allerdings führt die Erstellung und Integration von Simulationsmodellen sowie deren Nutzung immer noch zu erheblichen Kosten für Unternehmen. Seit geraumer Zeit werden Ansätze verfolgt, die ermöglichen sollen, Modelle effizienter zu erstellen und zu nutzen, aber erst die konsequente Nutzung eines Standards zur Datenmodellierung und zum Datenaustausch ermöglicht eine breitere Nutzung solcher Ansätze. In dieser Arbeit wird gezeigt, dass CMSD ein geeigneter Standard sowohl für die Erstellung von Simulationsmodellen als auch für deren Initialisierung ist.
Automatische Generierung adaptiver und lernfähiger Modelle zur Simulation von Produktionssystemen. - In: Tagungsband zum Doctoral Consortium der WI 2011, (2011), S. 9-16
In diesem Beitrag wird ein Ansatz zur automatischen Modelgenerierung und -adaption vorgestellt. Augenmerk liegt vor allem auf der Einsatzmöglichkeit im gesamten Produkt- bzw. Produktionslebenszyklus sowie bei der Integration in die betriebliche IT Infrastruktur, dazu wird auf standardisierten Datenaustausch mittels des Core Manufacturing Simulation Data (CMSD) Information Model gesetzt. Weitere Kernpunkte stellen die automatische Validierung und die Beschreibung bzw. Ermittlung von Steuerstrategien, z.B. Reihenfolgeregeln in Puffern, die Speicherung von manuell ergänzten Objekten/Verhalten und die Modellinitialisierung, dar. Die Beschreibung einer ersten prototypischen Implementierung einzelner Aspekte in Plant Simulation runden den Beitrag ab.
https://epub.uni-bayreuth.de/348/
Approximation of dispatching rules in manufacturing control using artificial neural networks. - In: 2011 IEEE Workshop on Principles of Advanced and Distributed Simulation, ISBN 978-1-4577-1365-1, (2011), insges. 8 S.
Automatic generation of simulation models has been a recurring topic in scientific papers for years. A common problem of all known model generation approaches is the generation of dynamic behavior, e.g. buffering or control strategies. This paper introduces a novel methodology for generation of dynamic behavior, based on artificial neural networks, which is usable directly in the simulation. We also test the approach in a manageable scenario; all results are illustrated via small simulation experiments.
http://dx.doi.org/10.1109/PADS.2011.5936774
Challenges for the automatic generation of simulation models for production systems. - Online-Ressource (PDF-Datei: 6 S., 366, 4 KB)CD-ROM-Ausg.: 2010 International Simulation Multiconference : Crowne Plaza Hotel, Downtown Ottawa, On., Canada, 12 - 14, 2010 / SCS; SISO. - [S.l.] : Omni Press, 2010. - ISBN 1-565-55344-6. - S. 545-549
http://www.db-thueringen.de/servlets/DocumentServlet?id=17445
Generierung und Integration von Simulationsmodellen unter Verwendung des Core Manufacturing Simulation Data (CMSD) Information Model. - In: Integrationsaspekte der Simulation: Technik, Organisation und Personal, (2010), S. 461-468
http://digbib.ubka.uni-karlsruhe.de/volltexte/1000019635
Modellgenerierung im Kontext der Digitalen Fabrik - Stand der Technik und Herausforderungen. - In: Integrationsaspekte der Simulation: Technik, Organisation und Personal, (2010), S. 37-44
http://digbib.ubka.uni-karlsruhe.de/volltexte/1000019635